open-llm-leaderboard-old/details_abacusai__bigyi-15b
收藏Hugging Face2024-05-04 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_abacusai__bigyi-15b
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对模型abacusai/bigyi-15b进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集从1次运行中创建,每次运行可以在每个配置中找到特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,results配置存储了所有运行的聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。
该数据集是在Open LLM Leaderboard上对模型abacusai/bigyi-15b进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集从1次运行中创建,每次运行可以在每个配置中找到特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,results配置存储了所有运行的聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集名称
- 名称: Evaluation run of abacusai/bigyi-15b
数据集组成
- 配置数量: 63个配置,每个配置对应一个评估任务。
- 创建来源: 数据集是在模型abacusai/bigyi-15b在Open LLM Leaderboard上的评估运行期间自动创建的。
- 运行次数: 数据集由1次运行创建,每个运行可以在每个配置中找到特定的分割,分割名称使用运行的时间戳。
- 最新结果: "train"分割始终指向最新的结果。
- 结果汇总: 一个额外的配置"results"存储所有运行的汇总结果,用于计算和显示在Open LLM Leaderboard上的汇总指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_abacusai__bigyi-15b", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳: 2024-03-10T12:57:48.733111
- 结果示例:
python
{
"all": {
"acc": 0.6368698545001125,
"acc_stderr": 0.03234846946993144,
"acc_norm": 0.6464868762775356,
"acc_norm_stderr": 0.03302252322767117,
"mc1": 0.2423500611995104,
"mc1_stderr": 0.015000674373570338,
"mc2": 0.37326698740277925,
"mc2_stderr": 0.01461558650400129
},
"harness|arc:challenge|25": {
"acc": 0.5324232081911263,
"acc_stderr": 0.014580637569995426,
"acc_norm": 0.560580204778157,
"acc_norm_stderr": 0.014503747823580127
},
"harness|hellaswag|10": {
"acc": 0.5749850627365066,
"acc_stderr": 0.004933349621589336,
"acc_norm": 0.7590121489743079,
"acc_norm_stderr": 0.004268088879039825
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.28,
"acc_stderr": 0.045126085985421276,
"acc_norm": 0.28,
"acc_norm_stderr": 0.045126085985421276
},
其他任务的结果...
}
配置详情
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_03_10T12_57_48.733111
- 路径: **/details_harness|arc:challenge|25_2024-03-10T12-57-48.733111.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-03-10T12-57-48.733111.parquet
- 分割: 2024_03_10T12_57_48.733111
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_03_10T12_57_48.733111
- 路径: **/details_harness|gsm8k|5_2024-03-10T12-57-48.733111.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-03-10T12-57-48.733111.parquet
- 分割: 2024_03_10T12_57_48.733111
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_03_10T12_57_48.733111
- 路径: **/details_harness|hellaswag|10_2024-03-10T12-57-48.733111.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-03-10T12-57-48.733111.parquet
- 分割: 2024_03_10T12_57_48.733111
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_03_10T12_57_48.733111
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-management|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-03-10T12-57-48.733111.parquet
- **/details_harness|hendrycksTest-professional_
- 路径:
- 分割: 2024_03_10T12_57_48.733111
- 数据文件:



